A Mathematical Method for Detecting LPI Signals
The unique characteristics of Low Probability of Intercept (LPI) waveforms, which are based on wideband specialized modulation techniques, LPI signals not only provide high resistance against electronic jamming but also significantly reduce the detectability of these signals, thereby establishing LPI signals as an effective electronic countermeasure. The Chirp Spread Spectrum (CSS) waveforms used in the LoRa protocol are a current example of signals possessing LPI characteristics. In complex electromagnetic environments where spectral density is high and the signal-to-noise ratio (SNR) drops to critical levels, the detection of these signals and the separation of their sub-characteristics remain an ongoing challenge in the literature. This study presents an original time-frequency analysis approach that enables the detection of CSS signals and parameter estimation with high accuracy under the aforementioned challenging environmental conditions. In tests conducted to validate the method’s performance, CSS signals with unknown parameters were subjected to different channel scenarios by generating spectral density in a laboratory environment. The proposed method was tested on IQ (In-Phase Quadrature) signal data recorded on a Software-Defined Radio (SDR) platform; using the developed algorithm, CSS signal parameters are extracted with high accuracy even under low SNR conditions where the signal power fell below the noise floor in some cases.